Congenital and Acquired Disorders of Platelet Function and Number

Does the Patient Have a Platelet-Related Bleeding Disorder?

Although our understanding of the mechanisms of platelet dysfunction and thrombocytopenia have improved significantly since Glanzmann's initial case report, the tests and skills required to recognize individuals with platelet-specific bleeding disorders have essentially remained unchanged. Typically most such individuals have their first encounter with a hematologist because of patient or physician concerns regarding either excessive mucocutaneous bleeding, thrombocytopenia, or a family history of bleeding. Frequently, a referring physician will already have performed several screening laboratory tests—including prothrombin time (PT), partial thromboplastin time (PTT), and fibrinogen level—to exclude other causes of bleeding. Sometimes other screening tests have been performed, such as bleeding time or testing with the platelet function analyzer (PFA-100, Siemens Healthcare Diagnostics, Tarrytown, NY). Therefore by the time of the initial evaluation by the hematologist—which will include history taking, physical examination, and careful evaluation of the blood smear—most of the critical information required for a correct diagnosis will already have been obtained.

The bleeding associated with platelet disorders primarily involves the skin and mucous membranes. Spontaneous joint or deep muscle bleeds are relatively uncommon in patients with platelet disorders. Bleeding manifestations within the skin can be characterized by their size, elevation, and distribution. Petechiae, purpura, ecchymoses, and hematomas can be found (see Chapter 10 ). Close attention should also be given to the distribution and elevation of the lesions, since this can be helpful in distinguishing bleeding related to platelet disorders from bleeding related to other causes, such as vasculitic disease or nonaccidental injury. For example, the lesions associated with Henoch-Schönlein purpura, a vasculitic disease, are raised (“palpable purpura”) and often appear in a classical distribution involving dependent surfaces and the lower extremities. Unusual or regular patterns of skin lesions (hand print, linear lesions, etc.) may suggest the possibility of nonaccidental injury.

Mucosal surfaces involved in platelet-related bleeding include the nasal membranes, as with epistaxis, bleeding within the oral cavity, upper and lower gastrointestinal hemorrhage, hematuria, and menorrhagia. In young women, menorrhagia is a common presenting complaint, and it has been estimated that up to 15% of women with menorrhagia have platelet dysfunction or von Willebrand disease (VWD). Platelet dysfunction should also be suspected in the presence of increased bleeding following dental extraction, tonsillectomy, or other surgical procedures.

Screening methods have been used in an attempt to identify individuals with a high likelihood of a platelet-related bleeding disorder or VWD. Availability of a rapid assay with high sensitivity would be expected to lower the need for further, more costly invasive testing. The bleeding time was the first of these methods to be widely used, but its clinical utility is questionable because of difficulties with standardization, reproducibility, and lack of sensitivity and specificity. The PFA-100 has also been proposed to have a role in the screening of individuals with suspected platelet dysfunction. The PFA-100 is a high shear stress–inducing device that simulates primary hemostasis by passing whole blood through an aperture cut into a membrane coated with collagen and either adenosine diphosphate (ADP) (50 µg) or epinephrine (10 µg). Platelets adhere to the collagen-coated surface and aggregate at the rim of the aperture. The platelet plug enlarges until it occludes the aperture, causing cessation of blood flow. The time to cessation of flow is recorded as closure time.

Initial studies focused on the efficacy of the PFA-100 in the evaluation of individuals with known VWD and severe platelet disorders like GT and Bernard-Soulier syndrome (BSS). In these patient populations with either VWD (von Willebrand factor [VWF] level of <30% of normal) or severe platelet disorders, both the sensitivity and specificity of PFA-100 testing approached 90%. These results suggest that the PFA-100 could be useful as an initial diagnostic tool in the evaluation of individuals with a suspected platelet-related bleeding disorder. More recently, however, the initial enthusiasm for the PFA-100 as a screening tool has diminished because of the device's reported low sensitivity (24% to 41%) in individuals with a mild platelet secretion defect or storage pool disorder.

Is the Defect Acquired or Congenital?

After the decision is made to pursue an investigation for platelet-related bleeding, a careful medical history taking, physical examination, and evaluation of the platelet count and blood smear can rapidly narrow the diagnostic possibilities. Several elements of the history provide critical information and should be explored in detail. History should be obtained about medication usage, particularly the use of aspirin or other nonsteroidal anti-inflammatory drugs (NSAIDs). Although NSAIDs are the most recognized drugs or group of drugs associated with platelet dysfunction, all medications should be suspect and evaluated, because the list of possible interactions is long and will continue to grow.

Close evaluation of the family and patient's medical history may provide clues about whether the bleeding disorder is acquired or congenital. Platelet dysfunction may be related to other systemic disorders. Uremia from renal impairment will cause platelet dysfunction. Cardiopulmonary bypass or extracorporeal membrane oxygenation will decrease the effectiveness of circulating platelets. Careful evaluation of the peripheral blood smear and blood count, as well as assessment for hepatosplenomegaly and lymphadenopathy, may reveal a need for further workup for myeloproliferative disease, malignancy, or aplastic anemia.

Additional elements of the history, physical examination, and peripheral smear are especially useful in differentiating immune thrombocytopenic purpura (ITP) from a congenital platelet disorder. The persistence of neonatal thrombocytopenia or a low platelet count despite the use of several standard therapies (intravenous immune globulin [IVIg], steroids, etc.) for ITP also points toward a congenital cause of the thrombocytopenia. Bleeding symptoms out of proportion to the platelet count should prompt consideration of a congenital platelet disorder. Patients with ITP typically have minimal bleeding until their platelet count decreases below 10,000 to 20,000/µL. Even at those low values, the presence of fresh reticulated platelets is usually enough to secure hemostasis. Bleeding observed at platelet counts of more than 30,000/µL suggests the presence of an underlying platelet dysfunction.

Many congenital platelet disorders are associated with other diseases, unique physical characteristics, or findings on the peripheral blood smear. For example, thrombocytopenia in the presence of auditory or renal dysfunction suggests an MYH9 -related thrombocytopenia, the presence of skeletal abnormalities may lead to the diagnosis of thrombocytopenia with absent radii (TAR) syndrome or amegakaryocytic thrombocytopenia with radioulnar synostosis, and the combination of albinism and platelet dysfunction suggests the diagnosis of Hermansky-Pudlak syndrome (HPS) (see later).

Medication-Related Platelet Dysfunction

Numerous medications have been associated with platelet dysfunction or thrombocytopenia. Inhibitors of platelet cyclooxygenase-1 (COX-1), aspirin, and other NSAIDs are the most commonly used drugs known to affect platelet function by decreasing the platelet production of the prostaglandin thromboxane A ₂ (TXA ₂ ). TXA ₂ is a potent secondary mediator of platelet activation, and its production within the platelet is mediated by the enzyme COX. Aspirin inhibits COX function by an irreversible covalent modification of the enzyme's active site; as little as 80 mg is required to completely inhibit TXA ₂ . Because of this irreversibility, the effectiveness of aspirin in inhibiting platelet function will persist until the platelets circulating at the time of administration are replaced (days). Other NSAIDs, however, act as reversible competitive inhibitors of platelet COX, and the duration of platelet inhibition will correlate with the half-life of the NSAID used (hours).

The thienopyridine ticlopidine and clopidogrel inhibit platelet activation by covalent modification of the P2Y ₁₂ receptor, a platelet receptor for ADP, the secondary mediator. Both of these agents are prodrugs that require cytochrome P-450–dependent pathways for their activation. As with aspirin, the effects of both ticlopidine and clopidogrel will persist throughout the platelet's entire life span.

Inhibitors of the fibrinogen receptor α _IIb β ₃ also block platelet function. The reversible competitive inhibitors tirofiban, eptifibatide, and abciximab are frequently administered in conjunction with vascular procedures, and all are rapidly cleared from the circulation following administration.

Several other drugs whose primary target is not the platelet may also cause platelet dysfunction. The penicillins and cephalosporins, which are β-lactam antibiotics, have been reported to affect platelet function. The mechanism of this inhibition is unclear, although it appears to be related to the effect of the drug on agonist–platelet receptor interaction. This effect typically occurs 2 to 3 days after the initiation of treatment. Drugs with primarily cardiovascular effects that may inhibit platelet function include the nitrates, the calcium channel blockers, and propranolol. The nitrates act as nitric oxide donors. Nitric oxide inhibits platelet function by decreasing platelet adhesion and recruitment to the forming thrombus. Among the psychotropic agents, the selective serotonin reuptake inhibitors (SSRIs) and tricyclic antidepressants may inhibit platelet function by decreasing the serotonin content of platelet-dense granules. Retrospective clinical studies suggest an increased risk of bleeding in patients receiving either SSRIs or tricyclics. More recently, it has been shown that altered serotonin transport in platelets affects α _IIb β ₃ activation and functional platelet responses. The consumption of alcohol has also long been associated with changes in platelet function. The impact of this effect appears to be sex dependent, with a greater effect on men. Both protective and detrimental effects on the cardiovascular system have been described, although moderate alcohol consumption seems to confer net protection against cardiovascular disease. Therefore, although it is clear that alcohol has defined effects on platelet activation and other components of the hemostatic and vascular systems, the extent of these effects and their clinical correlation with outcomes are less clear and need further investigation.

Cardiopulmonary Bypass and Extracorporeal Membrane Oxygenation

Cardiopulmonary bypass and extracorporeal membrane oxygenation both require that blood circulate over an artificial surface. Because the membrane lacks the anticoagulant properties of the endothelium, platelets can become activated. Platelet activation can result in both platelet dysfunction and increased platelet clearance. Increased platelet activation has been demonstrated in patients following cardiopulmonary bypass by using flow cytometry to evaluate platelet P-selectin expression and the formation of platelet-monocyte aggregates. The remaining circulating platelets also demonstrate decreased expression of the VWF receptor GPIb and the fibrinogen receptor α _IIb β ₃ , which suggests a possible explanation for platelet dysfunction. After cessation of cardiopulmonary bypass, normalization of the level of platelet receptors occurs gradually over 3 to 4 hours. It is not clear which laboratory tests of platelet function are useful to identify patients at excessive risk of bleeding or thrombosis following bypass. Because of the underlying platelet dysfunction, platelet transfusion is frequently employed in the case of hemorrhage after cardiopulmonary bypass even in the presence of an adequate platelet count (see Chapter 34 ). Patients on extracorporeal circulation or on a left ventricular assist device (LVAD) can also experience increased bleeding due to acquired von Willebrand syndrome, which can be diagnosed by demonstrating the loss of high-molecular-weight VWF multimers in plasma and should be treated with VWF replacement.

Uremia

Patients with chronic renal failure and uremia have long been recognized to be at increased risk of bleeding. The cause of the increased bleeding tendency is multifactorial and has been related to vascular abnormalities, anemia, and defects of platelet function and adhesion. Correction of the anemia of chronic renal failure with erythropoietin or red blood cell transfusion often results in normalization of abnormal bleeding times and improved platelet adhesion and aggregation.

In addition, multiple abnormalities of platelet function have been described in uremia . Among these are defects in platelet adhesion noted in various in vitro flow models using blood from uremic patients and decreased synthesis of TXA _2, the secondary mediator of platelet activation. Levels of many different circulating metabolites are increased in the uremic patient, and controversy remains regarding the identity of the substance(s) in uremic plasma responsible for the observed hemostatic abnormalities. One of the metabolites increased in uremic patients, guanidinosuccinic acid, can function as a nitric oxide donor and, in some studies, can reproduce several of the observed defects. Phenolic acid has also been proposed as a candidate agent, since it is present at concentrations within the plasma demonstrated to affect platelet function in vitro.

The treatment of bleeding in uremic patients can be challenging. Desmopressin (1-desamino-8- d -arginine vasopressin, or DDAVP) remains the agent of choice and has been demonstrated to effectively shorten the bleeding time in uremic patients. However, 4 to 8 hours after its administration, the bleeding time returns to baseline. Also, tachyphylaxis in response to repeated doses of DDAVP may occur. Improvement of the anemia by packed red cell transfusion may aid in hemostasis. Since the bleeding is presumed to be due to the presence of circulating metabolites, dialysis has been recommended in some circumstances. The benefit of dialysis, however, must be weighed against the potential hazards of anticoagulation and the exposure of platelets to an artificial membrane, with the potential for undesired platelet activation. Platelet transfusion is only transiently effective, since the transfused platelets are quickly inhibited in the uremic host. Recombinant factor VIIa (rFVIIa) has also been reported to effectively stop bleeding in uremic patients. The use of rFVIIa should be weighed carefully because of potential thrombotic complications.

Myeloproliferative Disorders

Both hemorrhagic and thrombotic complications have been reported in individuals with myeloproliferative disorders; however, the most common manifestation described is thrombosis. Hemorrhage can be problematic, and several possible reasons have been cited for the observed platelet defect. Megakaryocytic clonal abnormalities have been proposed to result in the production of defective platelets. Recently, the effect of JAK2 on the stability and localization of the thrombopoietin (TPO) receptor c-MPL has been demonstrated, which indicates a possible mechanism for the platelet abnormalities seen in these syndromes.

Abnormal in vivo platelet activation may result in the premature release of platelet granules, causing an acquired platelet defect. The loss of the highly hemostatic large VWF multimers has been associated with essential thrombocythemia, resulting in an acquired form of VWD. The concomitant use of antithrombotic agents increases the risk of bleeding in this patient population.

The use of platelet aggregation tests to define a set of patients with myeloproliferative disorders who are at risk of bleeding has limited clinical efficacy.

Hypersplenism

Hypersplenism refers to the thrombocytopenia that often occurs in individuals with splenic enlargement and that cannot be accounted for by other causes. Hypersplenism has been associated with a variety of pathologic conditions, including portal hypertension, lysosomal storage disorders, and myeloproliferative diseases. Increased sequestration of the circulating platelet population within the enlarged spleen and immunologic mechanisms have each been suggested as causes. In general platelet survival is normal, but up to 50% to 90% of the circulating platelets are sequestered in the large spleen. Thrombocytopenia associated with hypersplenism is usually mild (50,000 to 100,000 platelets per microliter) and requires no specific treatment. Severe thrombocytopenia or thrombocytopenia that interferes with a required therapy may require intervention, although this is rare. Different surgical approaches, including splenectomy, partial splenic embolization, and placement of a distal splenorenal shunt, have been used and may result in long-term normalization of the platelet count.

Congenital Platelet Disorders With a Normal Platelet Count

If a congenital platelet disorder is strongly suspected in an individual with a normal or near-normal platelet count, one of the first tests that should be performed is a platelet aggregation study. Such tests examine the ability of platelets to aggregate in stirred platelet-rich plasma in response to a panel of platelet agonists. Typical agonists evaluated are collagen, ADP, arachidonic acid, ristocetin, and epinephrine. Upon stimulation, platelets within the suspension become activated, change shape, release granular contents, and aggregate. All these events can be recorded by modern luminoaggregometry. When the turbid platelet-rich plasma is illuminated before the addition of an agonist, it scatters the transmitted light. The addition of the platelet agonist then results in an initial peak of increased transmission corresponding to a rapid change in platelet shape. Following this shape change the amount of light transmitted increases because of platelet clumping or aggregation. Two phases of aggregation can then be evaluated: a primary reversible aggregation phase caused by an immediate response of the platelets to exogenous agonists and a secondary irreversible phase corresponding to the response of the platelets to secondary mediators endogenously released by the platelets. Absence of or a large decrease in any of these components may be indicative of a particular congenital platelet disorder. Luminometry, used in combination with platelet aggregation, provides a sensitive evaluation of the release of adenosine triphosphate (ATP) by dense granules. ATP released by the platelets provides energy for the added light-producing enzyme luciferase, and a burst of light will be recorded. In patients with deficiency of a dense granules or a platelet release defect, this burst will be impaired. Without the platelets' own intrinsic “agonists” (i.e., ADP), often the secondary wave of platelet aggregation will be absent as well. Electron microscopy can then be used to distinguish a dense-granule deficiency from a platelet release defect. Closer evaluation of other platelet features by electron microscopy is often critically helpful as well.

Although luminoaggregometry and electron microscopy provide useful information about many structural and functional components of platelets, it should be emphasized that not all aspects of platelet activation are evaluated in these tests. Platelet function defects have been described in patients with a normal platelet count and a normal pattern of platelet aggregation, as in Scott syndrome. Also, platelet aggregation faces some of the same issues encountered with other platelet tests. It is accurate in diagnosing severe platelet disorders such as GT and BSS but seems to lack sensitivity and specificity for the milder platelet defects. Finally, since several common medications have an effect on platelet aggregation, it is imperative that patients avoid taking drugs known to affect platelet function, such as aspirin and other NSAIDs, or drinking alcohol at least 10 days before the test.